Reading battery health on a phone that has no battery health screen

Some phones print a battery health percentage in the settings. Most do not, and the ones that do rarely explain what the number is. Meanwhile every phone quietly records several things that are more informative than a single percentage, and a few of them are readable without any special access at all. Knowing which is which saves both money and an unnecessary battery replacement.

What wears out, and how

A lithium-ion cell loses capacity in two ways at once. Cycling — filling and emptying the cell — gradually consumes the lithium available for the reaction. Calendar ageing happens regardless of use, faster when the cell is hot and faster when it is kept at a high state of charge. A phone left in a drawer at full charge in a warm room ages measurably, without a single cycle.

Alongside capacity loss, internal resistance rises. That is the change users actually notice: a cell with high internal resistance sags in voltage under load, so the phone shuts down at what it believes is thirty per cent during a camera burst or a cold morning, then reports a healthy figure again once the load is removed. Sudden shutdowns at a plausible-looking charge level are the classic sign of resistance, not of lost capacity.

The numbers a phone already keeps

• Cycle count. One cycle is a full charge worth of energy, accumulated in fragments — two half charges make one cycle. Recent Android releases expose a cycle count and a manufacture date in the battery information page. A high count on a young phone tells you it has been worked hard; a low count on an old one tells you calendar ageing is the story.
• Design capacity against current capacity. The design figure is printed on the cell and in the phone's own data. The current figure is the fuel gauge's running estimate. The ratio is what a health percentage usually is.
• Charging voltage. Visible in the battery information screen on many builds. A cell that reaches its ceiling voltage very quickly and then tapers for a long time is telling you the usable capacity has shrunk.

Why the percentage is an estimate

No phone measures capacity directly. The fuel gauge infers it, by tracking current in and out and correcting against the voltage curve. That inference needs anchor points, which is why a gauge that has not seen a low state of charge for months can drift, and why a phone that has been kept between forty and sixty per cent for a long time sometimes reports odd figures until it has been run down once. That is the entire truth behind battery calibration: it does not restore capacity, it corrects the estimate.

A test that means something

Instead of chasing a percentage, measure runtime under a repeatable load. Charge to full, note the time, run the screen at a fixed brightness on a fixed task, and record the charge level each half hour. Repeat the same test a few months later. The comparison is yours, the conditions are controlled, and it captures both capacity loss and the resistance problem that a capacity figure misses entirely.

For a quick check of resistance alone, watch the charge level while the camera records video or while the phone navigates with the screen at full brightness. A cell in good condition falls steadily. A tired one drops in steps and sometimes jumps back up when the load stops.

What actually slows the wear

Three things, in order of effect: keep the cell cool, avoid leaving it at a very high charge for long periods, and avoid running it flat repeatedly. That is why charge limiters and adaptive overnight charging are worth leaving enabled, and why charging in a hot car is worse for a battery than any number of ordinary cycles. It also explains why fast charging is less damaging than its reputation suggests when the phone is cool, and more damaging than it looks when the phone is already warm — the mechanism is heat, not speed. The delivery side of that story is in why charging slows down.

When a replacement is the right answer

Replace the cell when runtime has fallen far enough to change how you use the phone, or when shutdowns happen at charge levels that should be comfortable. Both are judgements about use rather than thresholds on a screen. A phone that still lasts your day at eighty per cent of its original capacity does not need anything; a phone that shuts down at a quarter charge in the cold needs a cell regardless of what the health figure says.

If the phone is old enough that a replacement is a choice between the cell and a new handset, the software side matters too: a phone that no longer receives security patches is a different proposition from one that does, as we set out in the note on patch levels.